UC Davis Dermatology Online Journal

Title Quantification of ultraviolet (UV) radiation in the shade and in direct sunlight

Permalink https://escholarship.org/uc/item/4wc0f6tw

Journal Dermatology Online Journal, 25(7)

Authors Saric-Bosanac, Suzana S Clark, Ashley K Nguyen, Victoria et al.

Publication Date 2019

DOI 10.5070/D3257044801

License https://creativecommons.org/licenses/by-nc-nd/4.0/ 4.0

Peer reviewed

eScholarship.org Powered by the California Digital Library University of California Volume 25 Number 7| July 2019| Dermatology Online Journal || Original Article 25(7):4

Quantification of ultraviolet (UV) radiation in the shade and in direct sunlight

Suzana S Saric-Bosanac1, Ashley K Clark2, Victoria Nguyen3, Adrianne Pan4, Fang-Yi Chang5, Chin-Shang Li6, Raja K Sivamani1,4,7,8

Affiliations: 1Department of Dermatology, University of California, Davis, Sacramento, California, USA, 2Department of Dermatology, University of Pennsylvania, Philadelphia, Pennsylvania, USA, 3University of California, Davis, Davis, California, USA, 4College of Medicine, California Northstate University, Elk Grove, California, USA, 5Department of Statistics, University of California, Davis, Davis, California, USA, 6Department of Public Health Sciences, Division of Biostatistics, University of California, Davis, Sacramento, California, USA, 7Department of Biological Sciences, California State University, Sacramento, Sacramento, California, USA, 8Pacific Skin Institute, Sacramento, California, USA Corresponding Author: Raja Sivamani MD MS CAT, Department of Dermatology, University of California, Davis, 3301 C Street, Suite 1400, Sacramento, CA 95816, Tel: 916-703-5145, Email: [email protected]

Introduction Abstract Exposure to ultraviolet (UV) radiation may result in Ultraviolet (UV) radiation is associated with negative sunburn, skin damage, skin cancer, premature aging, health effects, including sun damage and skin cancer. damage to the eyes, and immunosuppression [1]. The purpose of this study is to compare the Long-term exposure to UV radiation is a major factor protective effects of the shade provided by a sun leading to the development of non-melanoma skin umbrella versus that provided by a tree. Sun sensors that register the level and dose of UV radiation were cancer. Malignant melanoma, although not as placed in the shade and in direct sunlight. common as non-melanoma skin cancer, is the major Measurements were recorded every half hour cause of skin cancer related death [1]. In the United between the hours of 12:30 p.m. and 3:00 p.m. in States, since the 1970s, the incidence of melanoma Sacramento, California. The results suggest that the has increased by 4% every year and many studies level of UV radiation in the shade is not zero. The suggest that exposure to UV radiation is correlated sensors located in tree shade indicated that over 5% with the risk of developing this type of cancer [1]. In of UV radiation was detected in the shade. The order to reduce the risk of skin cancer, individuals sensors located in sun-umbrella shade showed that should minimize their exposure to UV radiation [2]. greater than 17% of UV radiation reached the shade. When outdoors, the appropriate sun protective The sun sensors used in our study collected UV measures should be taken, including wearing sun radiation data relevant to UV index; however, they protective clothing, applying sunscreen, wearing did not differentiate between UVA, UVB, visible, and sunglasses, and staying in the shade when possible. infrared light. The amount of UV radiation detected in the shade is not zero, thus regular sunscreen use There are two forms of radiation caused by sunlight: and other sun protective practices should be direct radiation and diffuse radiation. Tree shade and followed to reduce the risk of sun damage and skin hats, for example, may protect the skin from direct cancer. radiation by blocking sunlight from reaching the skin. However, diffuse radiation may come from all directions and protective measures may not be as Keywords: ultraviolet, UV, radiation, skin cancer, sunlight, effective. For example, in tree shade there may be shade less direct radiation compared to that in direct

- 1 - Volume 25 Number 7| July 2019| Dermatology Online Journal || Original Article 25(7):4

sunlight, because trees are able to filter the direct accumulated UV dose according to the formula: component. A larger proportion of UV dose in the Dose = UVI × time (hours). The sun sensors also shade comes from diffuse radiation. provide a measurement of the UV level, which is an Even though diffuse radiation can be detected in the approximate measure of the theoretical UVI. Table 1 shade, utilization of shade is believed to be one shows how the theoretical UVI is related to the UV important strategy to minimize UV exposure and level measured by the sun sensors. protect the skin from UV damage [2]. A study of 50 The level display changes quickly depending on trees in Australia showed that the erythemal UV in direct sunlight exposure, whereas the dose reflects the shade depended on the tree canopy density and the cumulative UV exposure. The sun sensors were the height of the start of canopy above the ground, used to measure UV dose and level in the shade and with no significant dependence on canopy width or in direct sunlight. To further assess the role of shade, tree height [3]. Additionally, Parisi et al. showed that measurements were recorded under tree shade and on a summer afternoon, 61% of erythemal UV sun umbrella shade. radiation in tree shade was related to diffuse UV level and dose next to swimming pool component of UV [4]. Moreover, about 56% of Two swimming pools, one located in Davis, California ultraviolet radiation type A (UVA) radiation is due to (latitude 38.5) and the other in Sacramento, a diffuse component [4]. These values are similar to California (latitude 38.6) were used for data the percentage of diffuse UV radiation detected in collection. The measurements of UV dose and level full sun on a winter afternoon (diffuse erythemal UV, occurred at 30-minute intervals at the following 46%; diffuse UVA, 28%). The higher diffuse UV times (Pacific Standard Time): 12:30 p.m., 1:00 p.m., radiation in the summer tree shade may result in 1:30 p.m., 2:00 p.m., 2:30 p.m., and 3:00 p.m. The high UV exposure of skin and other parts of the body, measurements were done on a horizontal plane at including the eyes. Moreover, hats may provide more ground level in the shade that was provided by a protection from direct radiation than from diffuse sun-umbrella, as well as in direct sunlight (Figure 1). radiation [4]. The distance between coin dosimeters in the shade Quantitative measurements of UV radiation in and those in direct sunlight was approximately 12 various environments are needed for developing meters. The measurements were recorded on the optimal sun protection strategies. Although there same days at both locations. The theoretical UVI and have been numerous studies on quantifying UV the ambient temperature were the same at the two radiation in tree shade, few studies quantify UV locations, thus both locations were combined for the exposure in the shade near a swimming pool, a purposes of data analysis. At each time point, there potentially reflective surface. In this study, our were ten sensors in the shade (provided by the sun- primary objective was to quantify the UV exposure in umbrellas) and ten in direct sunlight, next to the direct sunlight by a swimming pool compared to the pool. The coin dosimeters used for data collection in UV exposure in the shade by a swimming pool direct sunlight were placed at a distance of one (shade provided by sun-umbrella). meter from the pool. All data were recorded on three sunny days in September 2016. Methods Table 1. UV Level Displays in Sun Sensor. Measurement of UV level and dose Number of “Level” Corresponding Intensity of bars displayed on theoretical UV UV measurements were conducted with a SunSense sunlight Coin®, which were kindly donated from SunSense sun sensor Index 1 Low 1-2 company (Norway). The sun sensors register 2 Moderate 3-5 erythema-weighted ultraviolet (UV) radiation 3 Strong 6-7 according to the international UV index (UVI). From 4 Very strong 8-10 the time the coin is turned on, it displays the 5 Extreme 11+

- 2 - Volume 25 Number 7| July 2019| Dermatology Online Journal || Original Article 25(7):4

and dose were measured on a horizontal plane at ground level in the tree shade and adjacent full sunlight using the sun sensors. The coins were moved throughout the day to maintain exposure to the sun, but they remained at an approximately equal distance from the tree trunk at all times (approximately 15 meters). The dosimeters that were recording data in the shade remained next to the tree trunk at all times. For each of the two trees, five coin dosimeters were collecting data in direct sunlight while five were in the shade, next to the tree trunk. Thus, at each time point, there were collectively ten measurements in the sun and ten in the shade. The data from the two trees were combined for the purposes of analysis as data collections were done on the same days, at the same times, at the same latitude. All data were recorded on three sunny days in September and October 2016. Statistical methods Since the UV dose is a continuous outcome variable, a linear mix-effects model was used to fit the Figure 1. The overall experimental setup, showing sun sensors in repeated measures UV dose data over time (12:30 tree shade (top panel) and sun umbrella shade (bottom panel). p.m. to 3:00 p.m.) to estimate the change rate in the UV dose for each of the two locations, direct sunlight UV level and dose next to tree and shade, and compare their change rate in the UV In Sacramento, California (latitude 38.6), two trees dose over time. For the UV level that is a 3-level (pine tree and chestnut oak tree) isolated from other ordinal categorical outcome variable, a random- trees and structures by at least five meters, were used effects proportional odds model was used to fit the in the study. The two trees were in different repeated measures UV level data over time (12:30 locations, but at the same latitude. The solar UV level p.m. to 3:00 p.m.) to estimate the change rate in the and dose in both tree shade and full sunlight were UV level for each of the two locations, direct sunlight measured. The measurements occurred at 30- and shade, and to compare their change rate in the minute intervals at the following times (Pacific UV level over time. A P-value <0.05 was considered Standard Time): 12:30 p.m., 1:00 p.m., 1:30 p.m., 2:00 statistically significant. All analyses were performed p.m., 2:30 p.m., and 3:00 p.m. The UV irradiation level with SAS v9.4 (SAS Institute Inc., Cary, NC, USA).

A B

Figure 2. Shade UV Level. Median UV level at each time point by sensor location in comparison to shade in A) sun umbrella shade and B) tree shade.

- 3 - Volume 25 Number 7| July 2019| Dermatology Online Journal || Original Article 25(7):4

A B Figure 3. Shade UV dose. UV Dose as a percent of dose received in direct sunlight in A) sun umbrella shade and B) tree shade.

Results UV dose (shade versus direct sunlight) UV level (shade versus direct sunlight) The UV dose detected in the shade was not zero. The The change in the UV level over time was not UV dose detected in sun umbrella shade increased statistically significant in the sun umbrella shade exponentially from 15% at baseline to 17.2% at group (Figure 2A). Similarly, the change in the UV 3:00p.m. (Figure 3A). The percent UV dose detected level over time was not statistically significant in the in the tree shade was 0% at baseline and increased direct sunlight group by the pool. The difference in to 5.6% at 3:00p.m. (Figure 3B). the change rate in the UV level between the two The difference in the change rate in the UV dose location groups (sun umbrella shade versus direct between the two location groups (sun umbrella sunlight) was not statistically significant (P=0.373). shade versus direct sunlight) was statistically The change in the UV level over time was not significant (P<0.0001) (Figure 4). Similarly, the statistically significant in the tree shade group difference in the change rate in the UV dose between (Figure 2B). However, the change in the UV level the two location groups (tree shade versus direct over time was statistically significant in the direct sunlight) was statistically significant (P<0.0001) sunlight group (P=0.024). The difference in the (Figure 5). change rate in the UV level between the two location groups (tree shade versus direct sunlight) was not Discussion statistically significant (P=0.9996). Our work demonstrates that shade is an effective mode of reducing UV radiation. However, it does not reduce UV radiation exposure to zero. In our study, we used a practical approach to our measurements. We utilized an umbrella near a pool, as this is how people typically seek shade when participating in activity near water and the area under a tree, since shade-seeking is a common approach used by people when outdoors. The sun sensor provides a reliable and reproducible measure of the UV dose received over time in the area where the coin was placed. It is important to note that the sun sensor readings may not match the meteorological (or theoretical) readings over time due to local conditions, such as buildings and trees (up to 50% of

Figure 4. Estimated trajectory of UV dose over time when UV radiation can be diffused), reflections, and comparing direct sunlight to poolside sun umbrella shade orientation of the sensor in relation to the sun.

- 4 - Volume 25 Number 7| July 2019| Dermatology Online Journal || Original Article 25(7):4

UV Reflectivity sun protective measures such as sunscreen, sun In our study, UV exposure in the shade provided by protective clothing, and hats, even while spending an umbrella near a pool is greater than that found time under a tree or umbrella. under a tree. This finding suggests that the type of The role of sunscreen in the shade shade matters (near a water source under an In addition to sunscreens and protective clothing, it umbrella versus not near water) owing to reflectivity is advised to seek shade for additional UV protection. of UV rays. The amount of UV radiation reaching People may assume spending time in the shade will earth’s surface varies widely. Several factors such as completely protect them from the sun. However, an cloud cover, the ozone, elevation, water depth and earth’s surface all account for the variation. individual will still receive UV radiation indirectly. Umbrellas provide partial UV protection. Their Sun At sea level, UVA comprises approximately 95% of Protection Factor (SPF) can range from 3-106 and the UV energy reaching the earth’s surface, whereas their Ultraviolet Protection Factor (UPF) up to 50+. ultraviolet light type B (UVB) comprises the Some studies have suggested that the amount of UV remaining 5% [5]. Water, such as the ocean’s surface, radiation that reaches beneath an umbrella can be can reflect about 5-8 percent of the UV radiation up to 84% of that in the sun [8]. However, our study reaching its surface with greater reflection in water demonstrated only 15-17% of the UV rays reached that is clearer [6]. Land can typically reflects 2-4 beneath the umbrella. There are multiple factors that percent of UV radiation [6]. Therefore, the large body may attribute to these differences. For example, our of water provided by a pool may reflect a large study design included a large beach umbrella amount of UV radiation contributing to the increased instead of a hand-held umbrella. Furthermore, our dose of UV exposure under an umbrella next to a umbrella was near a water source that scatters UV pool. rays in a different manner than concrete, grass, or Additionally, we found tree shade to be an effective sand [9]. mode of sun protection. Similar to our findings, other The importance of broad-spectrum sunscreen use is studies have concluded trees with large dense highlighted in our study. A greater proportion of UVA foliage are best at protecting from UV rays [7]. rays reach the earth’s surface in comparison to UVB However, as seen in our study, shade provided by a and UVA rays tend to scatter more than visible light. tree does not eliminate all UV exposure and there is Therefore, the UV index obtained by our sun sensors still risk of exposure. Therefore, it is important to use in the shade may reflect greater UVA exposure in the shade. Although the sensors used in this study do not differentiate between UVA and UVB, our findings warrant the development of future sensors that can differentiate UVA and UVB exposure. Limitations We were not able to eliminate all surrounding buildings from our collections. However, we utilized a direct sunlight control at each location to control for variations in exposures related to surrounding structures. Our study was limited to the Sacramento area and the UV exposures may be different in different geographic areas. Nevertheless, we utilized a control measurement for each shade measurement

Figure 5. Estimated trajectory of UV dose over time when and report the exposure in terms of percent of direct comparing direct sunlight to poolside tree shade. sunlight to normalize the data.

- 5 - Volume 25 Number 7| July 2019| Dermatology Online Journal || Original Article 25(7):4

Our study is limited to collection of UV light relevant a swimming pool was greater than 17%. The public to the UV index. This represents how current public should be aware that shade provides UV protection. health messages are developed and represents true However, individuals may still be at risk of skin UV exposure as a mix of UVA and UVB. The sensors damage related to UV exposure in the shade. were not able to differentially capture UVA, UVB, Individuals should follow sun protective measures, such as regular sunscreen application and sun visible light, and infrared light. As future sensors are protective clothing use, even in the shade, to reduce refined to subcategorize the radiation, future studies their exposure to ultraviolet radiation. will be able to differentiate and quantify the subgroups of radiation that are represented in the Potential conflicts of interest shade. RKS serves as a scientific advisor to LearnHealth and a consultant to Burt’s Bees, Dermala, and Tomorrow’s Conclusion Leaf. The other authors have no conflicts of interest Although tree shade and sun-umbrella shade to declare. provide protection from the damaging UV radiation, the amount of UV radiation detected in the shade is Acknowledgments not zero. Our study shows that the percentage of UV The project described was supported by the National radiation detected in tree shade was greater than Center for Advancing Translational Sciences, 5%. Similarly, the percentage of UV radiation National Institutes of Health, through grant #UL1 detected in the shade under a sun-umbrella next to TR001860 to AKC, and linked award TL1 TR001861.

References 1. WorldHealthOrganization. Ultraviolet radiation and the Am Acad Dermatol. 2011;64:748-58. [PMID: 21292345] INTERSUN Programme; Health effects of UV radiation. Available 6. Herman J, Celarier EA. Earth surface reflectivity climatology at from: http://www.who.int/uv/en/. Accessed on July 11, 2019. 340–380 nm from TOMS data. J Geophy Res-Atmos 2. Parisi AV, Turnbull DJ. Shade provision for UV minimization: a 1997;102(D23):28003-11. [DOI: 10.1029/97JD02074]. review. Photochem Photobiol. 2014;90:479-90. [PMID: 24417678]. 7. Parisi AV, Kimlin MG, Wong JC, Wilson M. Solar ultraviolet 3. Parisi AV. Effects of tree shade on solar ultraviolet exposures to exposures at ground level in tree shade during summer in south humans. UV radiation and its effects: an update. Conf, 26–28 east Queensland. Int J Environ Heal R. 2001;11:117-27. [PMID: March, Christchurch, New Zealand. 2002. 11382345]. https://eprints.usq.edu.au/1337/1/Parisi_2.pdf. 8. Turnbull DJ, Parisi A. Spectral UV in public shade settings. J 4. Parisi AV, Kimlin MG, Wong JC, Wilson M. Diffuse component of Photoch Photobio B 2003;69:13-9. [PMID: 12547492]. solar ultraviolet radiation in tree shade. J Photochem Photobiol B. 9. David HS. Epidemiological studies of sunlight and cataract: the 2000;54:116-20. [PMID: 1083654]. critical factor of ultraviolet exposure geometry. Ophthal 5. Sambandan DR, Ratner D. Sunscreens: An overview and update. J Epidemiol. 1994;1:107-19. [PMID: 8790617].

- 6 -